System Administration
cancel
Showing results for 
Search instead for 
Did you mean: 

ndd time outs still causing issues

Ratzie
Super Advisor

ndd time outs still causing issues

We have an app that opens up a telnet session to external sources.
Yes, I know this is an issue.

We are getting disconnects after 30 minutes, and I can not seem to find what the source is.

HP-UX B.11.11 U 9000/785

Here is our ndd settings:
-------------------------
tcp_time_wait_interval ----> 60000
tcp_conn_request_max ----> 4096
tcp_debug ----> 0
tcp_smallest_nonpriv_port ----> 1024
tcp_ip_abort_cinterval ----> 75000
tcp_ip_abort_interval ----> 600000
tcp_ip_notify_cinterval ----> 10000
tcp_ip_notify_interval ----> 10000
tcp_ip_ttl ----> 64
tcp_ip6_hop_limit ----> 64
tcp_keepalive_interval ----> 7200000
tcp_mss_def ----> 536
tcp_mss_max ----> 65495
tcp_mss_min ----> 120
tcp_naglim_def ----> 65535
tcp_old_urp_interpretation ----> 1
tcp_rexmit_interval_initial ----> 3000
tcp_rexmit_interval_max ----> 60000
tcp_rexmit_interval_min ----> 500
tcp_wroff_xtra ----> 32
tcp_deferred_ack_interval ----> 50
tcp_deferred_ack_max ----> 22
tcp_snd_lowat_fraction ----> 0
tcp_sth_rcv_hiwat ----> 0
tcp_sth_rcv_lowat ----> 0
tcp_dupack_fast_retransmit ----> 3
tcp_rwin_credit_pct ----> 50
tcp_rcv_push_wait ----> 16384
tcp_keepalives_kill ----> 1
tcp_smallest_anon_port ----> 49152
tcp_largest_anon_port ----> 65535
tcp_keepalive_detached_interval(read ----> name is non-existent for this module for a list of valid names, use name '?'
tcp_text_in_resets ----> 1
tcp_xmit_hiwater_def ----> 32768
tcp_xmit_lowater_def ----> 8192
tcp_recv_hiwater_def ----> 32768
tcp_xmit_hiwater_lfp ----> 65536
tcp_xmit_lowater_lfp ----> 16384
tcp_recv_hiwater_lfp ----> 65536
tcp_xmit_hiwater_lnp ----> 8192
tcp_xmit_lowater_lnp ----> 2048
tcp_recv_hiwater_lnp ----> 8192
tcp_rexmit_interval_initial_lnp(read ----> name is non-existent for this module for a list of valid names, use name '?'
tcp_syn_rcvd_max ----> 500
tcp_tw_cleanup_interval ----> 60000
tcp_do_conn_options ----> 1
tcp_fin_wait_2_timeout ----> 0
tcp_ts_enable ----> 2
tcp_sack_enable ----> 2
tcp_xmit_hiwater_max ----> 2147483647
tcp_recv_hiwater_max ----> 1073725440
tcp_cwnd_initial ----> 4
tcp_early_conn_ind ----> 1
tcp_spec_port ----> 80
tcp_smoothed_rtt ----> 1
tcp_delay_final_twh_ack ----> 1
tcp_discon ----> operation failed, Permission denied
tcp_discon_by_addr ----> operation failed, Permission denied
hatcp_enable ----> hatcp_enable_value 0
tcp_isn_passphrase ----> 0

3 REPLIES
Steven E. Protter
Exalted Contributor

Re: ndd time outs still causing issues

Shalom,

More information is needed than this.

There must be come kind of entry in /var/adm/syslog/syslog.log

SEP
Steven E Protter
Owner of ISN Corporation
http://isnamerica.com
http://hpuxconsulting.com
Sponsor: http://hpux.ws
Twitter: http://twitter.com/hpuxlinux
Founder http://newdatacloud.com
Ratzie
Super Advisor

Re: ndd time outs still causing issues

There is no really error in syslog. Just multiple entries every 5 minutes.

Apr 22 11:57:04 server inetd[3853]: registrar/tcp: Connection from snap (172.169.1.110) at Tue Apr 22 11:57:04 2008
Apr 22 12:02:04 server inetd[3878]: registrar/tcp: Connection from snap (172.168.1.110) at Tue Apr 22 12:02:04 2008

Here is the scenerio.
Users use exceed to pull back an Xwindows session app. With in the app it calls SwitchTerm to connect to a switch.
It's the switchTerm that times out, and we are in a "it's not us" thing between HP and SwitchTerm. I just want to confirm we have set everything correct on our end.
Highlighted
KehariSingh
Senior Member

Re: ndd time outs still causing issues

Hi

 

Please check in /etc/profile for TMOUT vlaue is it 1800sec (30min) or not. You can reduce this value to your wish.

The same can be found in the perticuler users profile @ its home location also.

 

 

Regards

Rishikesh Shukla